Global Options

R Markdown Cheatsheet

Help > Cheatsheets > R Markdown Cheatsheet

“Pandoc’s Markdown”

Markdown

Getting Started with Rmd files

Literate Programming: interweaving code, its output, and regular prose text

File > New File > R Markdown …

New Rmd File

The YAML header

---
title: 'Back to Basics'
subtitle: 'R you markDown?'
author: "Daniel Chen"
date: "June 26, 2017"
output: 
  revealjs::revealjs_presentation:
    theme: "night"
    highlight: "zenburn"
    css: styles.css
    self_contained: true
    reveal_options:
      slideNumber: true
      previewLinks: true
---

revealjs: http://rmarkdown.rstudio.com/revealjs_presentation_format.html

Output

Output Options

Main Document

R Chunks

Write:

```{r eval=TRUE}
n = 10
rnorm(n)
```

Output:

n = 10
rnorm(n)

##  [1] -0.64019032  0.60136155 -0.17928616  0.83450370  0.55648524
##  [6] -0.80748120  0.44146987 -0.01113623  0.80593002  1.52699277

Prose Text

The previous slide showed random draws from the normal distribution with mean = 0 and sd = 1.

There were 10 draws.

We can either

1. manually type in the number, n
2. print the value of n in an r chunk.

print(n)

## [1] 10

3. Or we can run R code inline as so: `r n`

Chunks

```{r}
# R code here
```

```{r chunk_name}
# R code here
```

```{r chunk_name, echo = TRUE}
# R code here
```

Documented, repliciable/reproducible, and automatic

The knit button (ctrl + shift + k)

rmarkdown::render(input,
                  output_format = NULL, output_file = NULL, output_dir = NULL,
                  output_options = NULL, intermediates_dir = NULL,
                  knit_root_dir = NULL,
                  runtime = c("auto", "static", "shiny", "shiny_prerendered"),
                  clean = TRUE,
                  params = NULL,
                  knit_meta = NULL, envir = parent.frame(),
                  run_pandoc = TRUE, quiet = FALSE, encoding = getOption("encoding"))

Parameters

DEMO 01-parameters

---
title: "parameters"
author: "Daniel Chen"
date: ""
output: html_document
params:
  data: ""
---

Parameterized Reports

Run in R/Rmd:

```{r}
rmarkdown::render(input = '01-parameters/params.Rmd',
                  quiet = TRUE,
                  params = list(
                      data = 'cars'
                  ))
```

Run in the command line:

1. install: pandoc pandoc-citeproc
2. shell:

Rscript -e "rmarkdown::render(input = '01-parameters/params.Rmd',
                              params = list(data = 'mtcars'))"

RMarkdown Website

• http://rmarkdown.rstudio.com/lesson-1.html

• notebooks
• slides
• interactive documents
• dashboards
• websites

• books

R Notebook

http://rmarkdown.rstudio.com/r_notebooks.html

• output: html_document -> output: html_notebook
  • output visible immediately beneath the input.
  • R Markdown documen:, all the code is sent to the console at once
  • Notebook: one line at a time is sent
  • Slightly better error handling
  • “knit” vs “previewed”
  • preview does not execute any of your R code chunks
    • it simply shows you a rendered copy of the markdown in your document along with the most recent chunk output
  • .nb.html

Other Languages (SQL)

DEMO 02-notebook

library(DBI)
library(RSQLite)

rm(list = ls())

con = dbConnect(SQLite(), dbname = "data/survey.db")

```{sql connection=con}
SELECT * from Person;
```

SELECT * from Person;

Data into R

```{sql connection=con, output.var=df_sql}
SELECT * from Person;
```

Data from R

name = 'William'

```{sql connection=con, output.var=df_sql}
SELECT * from Person WHERE personal = ?name;
```

Not much more different than r markdown/notebooks

• html: ioslides
• html: slidy
• html: revealjs (install.packages('revealjs'))
• pdf: beamer

output: 
  revealjs::revealjs_presentation

HTML works too!

Markdown documents can also render HTML!

It’s how I make a 2-column slide

.column-left2{
  float: left;
  width: 50%;
  text-align: left;
}
.column-right2{
  float: right;
  width: 50%;
  text-align: left;
}

<div class="column-left2">
  <center>
  Hello
  </center>
</div>

<div class="column-right2">
  <center>
  there!
  </center>
</div>

Intro to Shiny

DEMO 03-shiny_docs

• Introduce the concepts of Shiny without the boilerplate
• Minimal ui code
• server code is not in a function

Shiny Presentations

You can also have shiny apps run in a presentation like this!

No demo though.

flexdashboard

http://rmarkdown.rstudio.com/flexdashboard/

• Use R Markdown!
• Different layouts
  • row/column
  • storyboard
  • mobile

---
title: "Row Orientation"
output: 
  flexdashboard::flex_dashboard
---

Example 1

Example 2

Example 3

blogdown

• R Markdown and Hugo
• devtools::install_github('rstudio/blogdown')
  • blogdown::install_hugo()
  • blogdown::new_site()
    • RStudio projects
  • blogdown::serve_site()
• https://bookdown.org/yihui/blogdown/

bookdown

• documentation: https://bookdown.org/yihui/bookdown/

• books: https://bookdown.org/

• install.packages("bookdown")

• devtools::install_github("rstudio/bookdown")

Minimal Viable Product

DEMO 05-bookdown/01-minimal

• In separate rstudio project
• https://github.com/yihui/bookdown-minimal

---
title: "A Book"
author: "Frida Gomam"
site: bookdown::bookdown_site
documentclass: book
output:
  bookdown::gitbook: default
  bookdown::pdf_book: default
---

# Hello World

Hi.

Bye.

bookdown::render_book('index.Rmd', 'all')

rendered book will be under _book

bookdown-demo

• code: https://github.com/rstudio/bookdown-demo
• rendered: https://bookdown.org/yihui/bookdown-demo/

bookdown::render_book('index.Rmd', 'all')

M-K vs K-M

• “Merge and Knit” (M-K): default

• “Knit and Merge” (K-M): new_session = TRUE or new_session: yes

• K-M does not allow Rmd files to be in subdirectories, but M-K does.

Working directory

Notebooks

• chunk is always the directory containing the notebook .Rmd file.
• allows for same behaviour while knitting and in interactive mode
• get a warning if you try to change the working directory inside a notebook chunk

• revert back to the notebook’s directory once the chunk is finished executing

• use the knitr root.dir option
  • knitr::opts_knit$set(root.dir = normalizePath(".."))
  • only applies to chunks
  • relative paths in Markdown are still relative to the notebook’s parent folder

Working directory

RMarkdown

if (interactive()){
    data_dir <- './data'
} else {
    data_dir <- '../data'
}

feedback needed: https://github.com/chendaniely/rendR

Other things to look at/consider

• Get verbatim R chunks into R Markdown:
  http://rmarkdown.rstudio.com/articles_verbatim.html
• sweave and knitr:
  https://support.rstudio.com/hc/en-us/articles/200552056-Using-Sweave-and-knitr
• sweave child documents example:
  https://github.com/chendaniely/wicer-entice3

Thanks!

• @chendaniely
• https://github.com/chendaniely/2017-06-26-meetup-r_you_markDown